Thymomas alter the T-cell subset composition in the blood: a potential mechanism for thymoma-associated autoimmune disease.

Thymomas are the only tumors that are proven to generate mature T cells from immature precursors. It is unknown, however, whether intratumorous thymopoiesis has an impact on the peripheral T-cell pool and might thus be related to the high frequency of thymoma-associated myasthenia gravis. This study shows, using fluorescence-activated cell sorting-based analyses and T-cell proliferation assays, that thymopoiesis and T-cell function in thymomas correspond with immunologic alterations in the blood. Specifically, the proportion of circulating CD45RA(+)CD8(+) T cells is significantly increased in patients with thymoma compared with normal controls, in accordance with intratumorous T-cell development that is abnormally skewed toward the CD8(+) phenotype. Moreover, it is primarily the proportion of circulating CD45RA(+)CD8(+) T cells that decreases after thymectomy. The results also demonstrate that T cells reactive toward recombinant autoantigens are distributed equally between thymomas and blood, whereas T-cell responses to foreign antigen (ie, tetanus toxoid) are seen only among circulating T cells and not among thymoma-derived T cells. These functional studies support the hypothesis that thymopoiesis occurring within thymomas alters the peripheral T-cell repertoire. Because many thymomas are enriched with autoantigen-specific T cells, a disturbance of circulating T-cell subset composition by export of intratumorous T cells may contribute to paraneoplastic autoimmune disease arising in patients with thymoma. (Blood. 2000;96:3872-3879)

Neurofilament is an autoantigenic determinant in myasthenia gravis.

Intratumorous expression of a 153-kd protein (p153), which contains an acetylcholine receptor-like epitope, is the only tumor marker described to date that significantly associates with thymoma in paraneoplastic myasthenia gravis (MG). Here, we report that p153 is identical to the midsize neurofilament, as verified by immunohistochemistry, immunofluorescence, and western blot analysis. Furthermore, the acetylcholine receptor-like epitope of the midsize neurofilament (NF-M) was identified by peptide epitope mapping. We also show, using T-cell proliferation assays, a significantly increased response of intratumorous T cells to a recombinant midsize neurofilament fragment in thymoma patients with MG compared with MG patients with thymic follicular hyperplasia or thymoma patients without MG. The T cells of thymic follicular hyperplasia and thymoma patients without MG seem to be unresponsive to NF-M. In contrast, we found increased T-cell responses to recombinant acetylcholine receptor fragments in MG patients in general compared with non-MG patients. Increased T-cell responses to NF-M in patients with paraneoplastic MG might be the result of an abnormal positive selection of immature T cells within thymomas, caused by the expression of NF-M in neoplastic thymic epithelial cells. Our results offer further evidence that NF-M expression in thymomas is an autoantigenic determinant in MG.

Association of acetylcholine receptor alpha-subunit gene expression in mixed thymoma with myasthenia gravis.

OBJECTIVE: To investigate the association of MG with the transcription of muscular or neuronal acetylcholine receptor (AChR) subunit genes in thymomas. BACKGROUND: Many steps in the pathogenesis of MG have been elucidated but, with rare exceptions, its etiology is unknown. In patients with MG with thymoma, the tumor probably elicits autoimmunity to AChR, but it is enigmatic why MG develops in some patients but not in others. METHODS: Reverse transcriptase (RT)-PCR, immunohistochemistry, and immunofluorescence studies were carried out to investigate AChR expression in 35 patients with thymoma. Statistical analysis was used to specify significant differences between thymoma subtypes. RESULTS: Considering all thymomas (n = 35), no correlation was found between MG status and AChR gene expression as detected by RT-PCR. However, when histologically defined thymoma subtypes were studied separately, transcription of the muscular AChR P3A- alpha-subunit gene was significantly associated (alpha < 0.01) with the occurrence of MG in mixed thymomas (n = 17), but not in thymomas of the cortical type. For the other muscular AChR subunits (P3A+ alpha isoform, beta, gamma, delta, and epsilon) and the alpha2 and beta4 neuronal AChR subunits, no such correlation was detected. CONCLUSIONS: Expression of the P3A AChR alpha-subunit gene might be important for the pathogenesis of MG in mixed thymomas, suggesting etiologic heterogeneity of paraneoplastic MG among patients with histologically different thymoma subtypes.

Abnormal thymocyte development and generation of autoreactive T cells in mixed and cortical thymomas.

To gain insight into the pathogenesis of thymoma-associated myasthenia gravis, thymocyte maturation was investigated in mixed and cortical thymomas by three-color flow cytometry. Although we detected cells at all recognizable stages, we noted an unusual increased percentage of early CD4+/CD3- thymocytes--especially in mixed thymoma--and a pronounced decreased percentage of mature CD4+/CD3+ cells in cortical thymomas as well. The percentage of CD3+/CD69+ cells that arose after positive selection was reduced in both thymoma subtypes compared with control thymuses, which suggests differences in the rate or efficiency of positive selection particularly in mixed thymomas. Mature T cells in 10 of 11 thymomas were not activated in situ as shown by the absence of CD25 expression. After stimulation with recombinant human acetylcholine receptor alpha-subunit fragments, thymocytes from 8 of 11 thymomas of both subtypes proliferated more strongly than those from controls, regardless of whether the donors were myasthenic. Responses of residual thymus cells to tetanus toxoid correlated well with those of autologous blood T cells, whereas those from the thymomas clearly did not--implying minimal colonization of thymomas by mature recirculating T cells. In conclusion, our results show that cortical and mixed thymomas exhibited differences in thymocyte maturation. Nevertheless, both thymoma subtypes seem to contribute to the pathogenesis of paraneoplastic myasthenia gravis by generating naive but potentially autoaggressive T cells; in some thymomas, these cells may then be actively immunized inside the tumor.